Hmm, I'm not sure about this one. For example, what happens if something takes an abstract method, and then, via the SPI, changes the method or class to not be abstract (but doesn't provide any implementation of the method/class). In this they will get a runtime error when we try to call the method or instantiate the class. On 31 Oct 2009, at 01:56, Gavin King wrote: > I suppose that there are places where the container checks if a class > or method is abstract. I guess it makes sense to indirect that? > > -- > Gavin King > gavin.king(a)gmail.com > http://in.relation.to/Bloggers/Gavin > http://hibernate.org > http://seamframework.org > _______________________________________________ > weld-dev mailing list > weld-dev(a)lists.jboss.org > https://lists.jboss.org/mailman/listinfo/weld-dev

Ok, I see. Personally, I think this could wait until the next release, unless there is a good use case where this would be required. On 1 Nov 2009, at 14:39, Gavin King wrote: > Yeah that was my initial thought too. But what if, for example, you > overrode isAbstract() and then also wrapped the Producer? > > Sent from my iPhone > > On Nov 1, 2009, at 9:11 AM, Pete Muir <pmuir(a)redhat.com&gt; wrote: > >> Hmm, I'm not sure about this one. For example, what happens if >> something takes an abstract method, and then, via the SPI, changes >> the method or class to not be abstract (but doesn't provide any >> implementation of the method/class). In this they will get a runtime >> error when we try to call the method or instantiate the class. >> >> On 31 Oct 2009, at 01:56, Gavin King wrote: >> >>> I suppose that there are places where the container checks if a class >>> or method is abstract. I guess it makes sense to indirect that? >>> >>> -- >>> Gavin King >>> gavin.king(a)gmail.com >>> http://in.relation.to/Bloggers/Gavin >>> http://hibernate.org >>> http://seamframework.org >>> _______________________________________________ >>> weld-dev mailing list >>> weld-dev(a)lists.jboss.org >>> https://lists.jboss.org/mailman/listinfo/weld-dev >> > _______________________________________________ > weld-dev mailing list > weld-dev(a)lists.jboss.org > https://lists.jboss.org/mailman/listinfo/weld-dev

Ah. My main concern is that it would be very easy for users to break stuff totally. I think the spec would need to define some validation. On 1 Nov 2009, at 15:30, Gavin King wrote: > Well, yes, but it is very difficult to add stuff to interfaces which > are implemented by users. That's a breaking change.... > > On Sun, Nov 1, 2009 at 10:41 AM, Pete Muir <pmuir(a)redhat.com&gt; wrote: >> >> Ok, I see. >> >> Personally, I think this could wait until the next release, unless there >> is >> a good use case where this would be required. >> >> On 1 Nov 2009, at 14:39, Gavin King wrote: >> >>> Yeah that was my initial thought too. But what if, for example, you >>> overrode isAbstract() and then also wrapped the Producer? >>> >>> Sent from my iPhone >>> >>> On Nov 1, 2009, at 9:11 AM, Pete Muir <pmuir(a)redhat.com&gt; wrote: >>> >>>> Hmm, I'm not sure about this one. For example, what happens if >>>> something takes an abstract method, and then, via the SPI, changes >>>> the method or class to not be abstract (but doesn't provide any >>>> implementation of the method/class). In this they will get a runtime >>>> error when we try to call the method or instantiate the class. >>>> >>>> On 31 Oct 2009, at 01:56, Gavin King wrote: >>>> >>>>> I suppose that there are places where the container checks if a class >>>>> or method is abstract. I guess it makes sense to indirect that? >>>>> >>>>> -- >>>>> Gavin King >>>>> gavin.king(a)gmail.com >>>>> http://in.relation.to/Bloggers/Gavin >>>>> http://hibernate.org >>>>> http://seamframework.org >>>>> _______________________________________________ >>>>> weld-dev mailing list >>>>> weld-dev(a)lists.jboss.org >>>>> https://lists.jboss.org/mailman/listinfo/weld-dev >>>> >>> _______________________________________________ >>> weld-dev mailing list >>> weld-dev(a)lists.jboss.org >>> https://lists.jboss.org/mailman/listinfo/weld-dev >> >> > > > > -- > Gavin King > gavin.king(a)gmail.com > http://in.relation.to/Bloggers/Gavin > http://hibernate.org > http://seamframework.org

* To determine class, method, field, constructor accessibility * To determine if a class, method, field, constructor is final * To determine if a class,, method, field, constructor is static * To determine if a method, field, constructor is transient * To determine if a class is abstract Do you want info on which bits of the lifecycle use this info? On 1 Nov 2009, at 15:38, Gavin King wrote: > In what circumstances does the RI currently call getModifiers() for > Methods and Classes? Can you make a list? > > On Sun, Nov 1, 2009 at 11:35 AM, Pete Muir <pmuir(a)redhat.com&gt; wrote: >> Ah. >> >> My main concern is that it would be very easy for users to break >> stuff >> totally. I think the spec would need to define some validation. >> >> On 1 Nov 2009, at 15:30, Gavin King wrote: >> >>> Well, yes, but it is very difficult to add stuff to interfaces >>> which >>> are implemented by users. That's a breaking change.... >>> >>> On Sun, Nov 1, 2009 at 10:41 AM, Pete Muir <pmuir(a)redhat.com&gt; >>> wrote: >>>> >>>> Ok, I see. >>>> >>>> Personally, I think this could wait until the next release, >>>> unless there >>>> is >>>> a good use case where this would be required. >>>> >>>> On 1 Nov 2009, at 14:39, Gavin King wrote: >>>> >>>>> Yeah that was my initial thought too. But what if, for example, >>>>> you >>>>> overrode isAbstract() and then also wrapped the Producer? >>>>> >>>>> Sent from my iPhone >>>>> >>>>> On Nov 1, 2009, at 9:11 AM, Pete Muir <pmuir(a)redhat.com&gt; wrote: >>>>> >>>>>> Hmm, I'm not sure about this one. For example, what happens if >>>>>> something takes an abstract method, and then, via the SPI, >>>>>> changes >>>>>> the method or class to not be abstract (but doesn't provide any >>>>>> implementation of the method/class). In this they will get a >>>>>> runtime >>>>>> error when we try to call the method or instantiate the class. >>>>>> >>>>>> On 31 Oct 2009, at 01:56, Gavin King wrote: >>>>>> >>>>>>> I suppose that there are places where the container checks if >>>>>>> a class >>>>>>> or method is abstract. I guess it makes sense to indirect that? >>>>>>> >>>>>>> -- >>>>>>> Gavin King >>>>>>> gavin.king(a)gmail.com >>>>>>> http://in.relation.to/Bloggers/Gavin >>>>>>> http://hibernate.org >>>>>>> http://seamframework.org >>>>>>> _______________________________________________ >>>>>>> weld-dev mailing list >>>>>>> weld-dev(a)lists.jboss.org >>>>>>> https://lists.jboss.org/mailman/listinfo/weld-dev >>>>>> >>>>> _______________________________________________ >>>>> weld-dev mailing list >>>>> weld-dev(a)lists.jboss.org >>>>> https://lists.jboss.org/mailman/listinfo/weld-dev >>>> >>>> >>> >>> >>> >>> -- >>> Gavin King >>> gavin.king(a)gmail.com >>> http://in.relation.to/Bloggers/Gavin >>> http://hibernate.org >>> http://seamframework.org >> >> > > > > -- > Gavin King > gavin.king(a)gmail.com > http://in.relation.to/Bloggers/Gavin > http://hibernate.org > http://seamframework.org > _______________________________________________ > weld-dev mailing list > weld-dev(a)lists.jboss.org > https://lists.jboss.org/mailman/listinfo/weld-dev